Claybook [XO, PS4, PC, NX]

I'm à YouTuber where is my key ? :LOL:

Really lovely game. Still no VR experimentation yet ?
Sent you a private message with a Claybook key.

We have Oculus Rift at office. Will do some testing during the early access period, but now we are focusing 100% on the launch.
 
Sent you a private message with a Claybook key.

We have Oculus Rift at office. Will do some testing during the early access period, but now we are focusing 100% on the launch.

Wow thanks That's so kind of you ! I Was kidding but now i cannot wait to get back from Work !
 
I want some multiplayer footage!
Some coop splitscreen footage (Italian):

This video is from a bit older version. We did lots of small level improvements based on watching Youtuber videos and their mistakes. Added new tutorial messages, etc for features that weren't properly introduced earlier.

All levels can be played in cooperative mode. Most levels work perfectly, but we still need to add some alternative routes to make some levels a bit more enjoyable with more than one player.
 
Last edited:
Great job sebbbi. You've done something very original here. I've never seen a game this mouldable/deformable, and so organic at that. Seeing the environment bend and dent with contact from the player as it also gets collor from the surfaces it touches other things is very satisfying. It looks very un-gamey because of how soft and organic it is. good work man.
 
tenor.gif
 
just bought. congrats sebbbi! I know these types of things is a tough journey for family while you're trying to get the product out the door. Wish you the best of success.
 
I am curious about the rewind feature. Is this feature exact? Exact in the sense that you really back track the exact way (plus movement). Or is does it simulate the movement backwards, i.e. there are very small differences as I would assume a simulation is not machine precision exact when going back and forth...?
 
I am curious about the rewind feature. Is this feature exact? Exact in the sense that you really back track the exact way (plus movement). Or is does it simulate the movement backwards, i.e. there are very small differences as I would assume a simulation is not machine precision exact when going back and forth...?
There's a separate rewind history for each shape. It tracks only position and rotation and morph state (blend between shapes). Deform state is not tracked, but it is maintained (each shape is formed out of 16 thousand particles, recording that wouldn't be possible). When you rewind, you only rewind back one shape at a time. Other shapes and world state is not rewound. This is a bit like Braid. Because each shape has their own history, you can do some tricks with the rewind mechanic. We don't have any puzzles in early access launch version that require use of the rewind, but you can certainly get faster time on many puzzles by using it creatively. Final version will have much more levels, so expect to see some that require rewind tricks.
 
How big could you make the world if everything on it was particle based like the interective shapes?
 
How big could you make the world if everything on it was particle based like the interective shapes?
3d is much harder than 2d. Imagine 2k*2k 2d world. That's roughly two screens (scrolling) at 1920x1080. Not a big level. 4 million pixels in total. Now scale this to 3d, and you get 2k*2k*2k. Which requires two thousand times more memory and computational performance. And that's not actually even true, since 2d neighborhood operations in grid access 3x3-1 = 8 neighborhood cells, while the same operations in 3d access 3x3x3-1 = 26 neighborhood cells. Which is over 3.25x more expensive (in addition to the two thousand time increase in N). There's also a problem that 3d allows you to zoom close to any surface. So you actually need even more resolution than this to present a pixel perfect game from every possible camera angle (similar to the 2d version).

But fortunately there are tricks. You can sleep the particles when they settle down. If most movement is near surfaces, then the computational cost scales roughly by n^2 instead of n^3. This obviously assumes that majority of the volume area is either deep inside the ground or air. This is often true. So you (and me) can still keep our dreams alive about a future sandbox simulation game (with actual physically plausible sand). I have thought about it :)
 
And bought. Have just started reading about SDF and volumetric rendering so I can count this as research. Now just have to get through a day at work battling MIFID regulatory work and xml :)
 
Back
Top